Stellarlinkco/codex is a Rust-first coding agent, forked from openai/codex, designed for advanced AI-driven coding workflows. It targets developers, researchers, and power users needing robust multi-model support, multi-agent collaboration, and long-running orchestration. The project aims to provide an open-source alternative with features inspired by Claude Code, enhancing productivity via agent teams and a web interface.

How It Works

This project adopts a Rust-first architecture, extending openai/codex with a focus on sophisticated agent interactions. It implements Claude Code-style workflows, enabling agent teams, custom hooks, and direct Anthropic API integration. A key design choice is the Web UI, served by codex serve, centralizing control. The architecture prioritizes multi-model support for collaboration between different AI models and supports long-running orchestration for complex, continuous tasks.

Quick Start & Requirements

Installation is streamlined via a shell script for macOS, Linux, and WSL (curl -fsSL https://raw.githubusercontent.com/stellarlinkco/codex/main/scripts/install.sh | bash) or PowerShell for Windows (irm https://raw.githubusercontent.com/stellarlinkco/codex/main/scripts/install.ps1 | iex), placing the binary in ~/.local/bin. Linux users require glibc >= 2.35 (e.g., Ubuntu 22.04+). Further details are available in the Docs, Contributing, and Installation & Building guides.

Highlighted Details

• A Rust-first fork of openai/codex focusing on advanced agent capabilities.
• Implements Claude Code-style workflows: agent teams, hooks, Anthropic API integration.
• Features a dedicated Web UI accessible via codex serve.
• Designed for multi-model support, multi-agent collaboration, and long-running orchestration.

Maintenance & Community

The project acknowledges sponsorship from PackyCode, an API relay service provider, offering a discount to users. Community engagement is encouraged via the Contributing guide.

Licensing & Compatibility

The repository is licensed under the AGPL-3.0 License. This strong copyleft license mandates that derivative works must also be distributed under AGPL-3.0, potentially posing compatibility challenges for integration into closed-source commercial products.

Limitations & Caveats

As a fork integrating complex features like multi-agent collaboration and long-running orchestration, the project may still be under active development. Specific limitations, known bugs, or performance caveats are not detailed in the provided README snippet.